Have retail VIX Exchange Traded Products (ETPs) like the iPath S&P 500 VIX Short-Term Futures ETN (VXX), VelocityShares Daily Inverse VIX Short-Term ETN (XIV), ProShares Short VIX Short-Term Futures ETF (SVXY), etc. become large enough to have a meaningful impact on the broader US equity market? This is a question I posed in a recent Wall Street Journal blog and one Vance Harwood and I have been discussing for some time now. The answer, and more importantly the reasons behind it, might surprise you.

The traditional argument goes something like this. VIX products are derivatives of the S&P 500 options market, the VIX being a measure of its 30-day implied volatility, the VIX futures a tradable reflection of where the value of the VIX might be in the future, and the VIX ETPs a portfolio of VIX futures.  As a derivative of the S&P 500 options market, many believe these retail ETPs that merely reference the much larger wholesale market for S&P options would have little impact on it. This was certainly my thinking in the mid-2000s during our early development of the first VIX Exchange Traded Note, the VXX.

However, with the growth of interest in VIX ETPs, and particularly the growth in interest in the short volatility products, this argument may no longer hold much water. Let me explain.

To start with, the absolute size of the products is not as important as who holds them and what they do with them. Two retail investors with opposing views on the future of volatility might buy and sell a unit of VXX between themselves effectively creating a contract for difference that has no impact on the greater volatility market. If VXX rises, the investor who is long profits while the investor who is short losses. The net impact to the volatility market of this type of position is zero, which is to say that no net exposure is passed on into the S&P options market.

However, not all ETP trades offset in this way, and there is a growing consensus in the volatility market that the overwhelming retail interest in VIX ETPs is on the short volatility side, i.e short the long products like VXX, and long the inverse products like XIV.  This imbalance may be due to investors predicting a decline in equity volatility, but is just as likely to be the result of recently strong returns from strategies looking to harvest the volatility risk premium. See my earlier article Making Volatility Investable.

If this is the case, and the end users of volatility products, i.e those seeking to take a position rather than hedging it with other volatility products, are net short, it is likely that professional market makers and dealers are net long the other side of the trade. But unlike end users, market makers and dealers are subject to strict risk limits and make their money from flow and not position taking. As a result, the professionals hedge their volatility exposure with whatever products suit best them, including VIX futures, S&P options and Variance Swaps. Because VIX futures and S&P 500 variance swaps (see my previous article on this subject) are effectively just contracts for difference and merely move the risk from one trader’s book to another, ultimately all of these hedges find their way back to the S&P options market.

Now, this is where things get interesting. If dealers are long volatility in VIX ETPs (as a result of retail customers being net short), and dealers are short S&P options as a hedge, an interesting dynamic is created that may one day exacerbate moves in the broader equity markets.

So how might this dynamic contribute to a rise in volatility? The answer lies in the differing dynamics of VIX ETPs and S&P 500 Index options. Delta hedged options used to hedge VIX ETPs need to be re-hedged each time the S&P Index moves up or down – this is know in the derivatives world as Gamma hedging, and refers the amount of additional Delta (stocks and S&P futures) that needs to be bought (or sold) when the Index moves 1%. For example: $1m of Gamma indicates that if the S&P Index was to move 1% in either direction, the trader would need to buy (or sell) $1m worth of stocks and futures. If the trader is long options, they will be long Gamma meaning each time the market rallies they will be selling delta, and each time it falls they will be buying delta – a comfortable position to be in, selling high and buying low. However, the opposite applies when the trader is short options and short Gamma. Each time the market rallies the trader is forced to re-hedge by buying stocks and futures, and each time it falls they have to sell those stocks and futures. This can be uncomfortable and, in a volatile market, catastrophic!

So professional dealers and market makers and the hedge funds they lay it off to are probably long VIX ETPs and short options, leaving them in a short Gamma position as described above, the size of which is related to the net position held by unhedged retail investors, rather than the absolute size of the VIX ETP. Remember the zero effect of my contract for difference example earlier.

So how big could this short gamma position be, and could it be big enough to have an impact on the broader equity market? Well yes is the answer, and growing interest in short volatility products by retail investors is driving its size each day.

Presently there is more than $3.5bn of VIX ETP AUM out there, the largest being VXX at $1.1bn, XIV $840m, SVXY $575m, the ProShares Ultra VIX Short-Term Futures ETF (NYSEARCA:UVXY) $380m, and VelocityShares Daily 2x VIX Short-Term ETN (Nasdaq:TVIX) $200m, (the last two being twice leveraged so have double the volatility exposure) [Yahoo Finance August 28, 2017]. The dollar exposure to volatility that this AUM represents (S&P 500 Vega) amounts to about $300m (calculated by dividing the total AUM by the present VIX Level).

A reasonable proxy hedge for these products might be a strip of 60-day S&P 500 options, or to really simplify things, a 60 day at the money straddle. I use 60-day options as my proxy because the average maturity of the VIX futures held in a VIX ETP’s portfolio is about 30 days, and the VIX itself references 30-day volatility beyond that.

And this is where things start getting interesting. A $300m volatility exposure in 60-day options amounts to almost $20bn of ATM option gamma (readers should be able to calculate this themselves using an online option pricer or a Bloomberg terminal, and, for the sake of this example, I have used a 12% vol and 50 delta puts and calls).

Now, remember what a short $20bn gamma position translates to: for every 1% the S&P 500 moves down, the group of market makers, dealers, and hedge funds that hold this hedge will typically have to sell approximately $20bn of S&P 500 stock and futures to rebalance their VIX ETP position!

The most liquid hedge for the broad US equity market is the CME’s S&P 500 E-Mini future that has recently been trading with good volume of around 1,200,000 contracts per day. The multiplier on the contract is 50 and the most liquid September contract is trading just at a value just over 2,400. This amounts to about $140bn dollars of S&P 500 delta traded each day.

So for a 1% move in the S&P 500 market, VIX ETP hedging could account for as much as a seventh of all the day’s volume and for moves larger than 1% it may account for significantly more. This is a staggering observation given the continued growth in short volatility interest in VIX ETPs. If short volatility interest continues to grow, the volatility of the S&P 500 market may actually rise, as dealers rush to cover their short Gamma positions. Interestingly though, this rise in volatility may cause the short volatility ETP trade to become less profitable, and may ultimately restore some balance to the retail volatility market.